To investigate this, we compare the lending practices of a pioneering big tech lender, MyBank, with a major retail bank in China, Bank X. The study highlights stark differences in various aspects of big tech lending compared to Bank X’s practices. Big tech borrowers tend to have more restricted credit access, with a significant majority (81%) receiving their first business loan from a big tech lending program. These loans are typically uncollateralized, with smaller credit limits and loan sizes but carry higher interest rates.

Delinquency rates are higher for big tech loans overall, largely due to borrowers without a repayment history, accounting for nearly half of these loans. However, among borrowers with a repayment record, the delinquency rate is notably lower (1.2%), resembling Bank X’s loan performance.

Interestingly, even during the COVID-19 pandemic, big tech lending remained stable and, in some cases, the delinquency rates slightly decreased compared to Bank X’s loans. The study shows that the fast repayment of big tech loans—often settled well before their maturity—suggests that borrowers primarily use these loans for short-term liquidity needs, limiting the lender’s exposure to long-term risks.

Analyzing a subset of borrowers who accessed both big tech and Bank X’s loans, it’s surprising that despite their better credit quality, they still received similarly high interest rates and low credit limits from the big tech lender. This indicates that big tech lending doesn’t directly compete for higher quality borrowers. Fast repayment patterns persist in this subset, suggesting a consistent trend in using big tech loans for short-term liquidity needs.

Moreover, the analysis reveals that big tech lenders attract borrowers with short-term liquidity needs by providing convenience along with high interest rates. These borrowers, even with cheaper and sufficient credit lines from Bank X, opt for big tech loans, showcasing the advantages of big tech lending beyond financial terms.

In the case of overlapped borrowers, big tech loans exhibit lower delinquency rates than Bank X’s loans, validating the information advantage of the big tech lender and explaining the better risk performance of these loans.

In conclusion, the study presents a nuanced perspective on big tech lending, highlighting its unique ability to serve borrowers overlooked by traditional banks. Big tech lending primarily caters to short-term liquidity needs within its ecosystem rather than directly competing with banks for a complete spectrum of credit services. This model doesn’t aim to monopolize all lending but rather focuses on specific underserved segments, leveraging the tech company’s strengths in data and technology.

All three bank failures in March and May 2023 involved banks with low branch density. Their national rankings by total assets, total deposits, and the number of branches in 2022 are detailed in Table 1. Their branch densities were notably below the 10th percentile of the branch density distribution in 2022.

In our research paper, “Bank Branch Density and Bank Runs,” we demonstrate that these three bank failures were part of a larger issue, and a general trend links low branch density with increased deposit instability. Digital banking allows banks with lower branch density to swiftly grow and entice depositors during stable times (see figure 1). However, as economic conditions worsened and interest rates rose, these significant inflows of deposits turned into “hot money” and shifted course.

Our analysis unveils a clear and statistically significant correlation between bank branch density and stock returns during these bank failures. Around the collapse of Silicon Valley Bank, a one standard deviation decrease in branch density correlates with 4 percentage points lower returns (see figure 2). Similarly, during the failure of First Republic Bank, a one standard deviation decrease in branch density is linked to 1.4 percentage points lower returns. This implies that lower branch density is associated with notably reduced stock returns during uncertain times.

These lower stock returns likely mirror investor concerns regarding the financial stability of banks with low branch density. We also find that branch density had a positive correlation with deposit flows in the first quarter of 2023. A one standard deviation decrease in branch density relates to a 4.4 percent net outflow of uninsured deposits. Some banks with low branch density continued to attract insured deposits even in 2023, resulting in insignificant changes in their total deposits. However, among banks that witnessed net deposit outflows, those with low branch density suffered more substantial outflows.

One potential explanation for the underperformance of banks with low branch density in 2023 revolves around their depositors. Banks with low branch density primarily attract uninsured deposits through digital banking, which appeals to corporations and tech-savvy households with substantial funds to deposit. We reveal that banks that heavily invested in information technology (IT) in the past had lower branch density in 2022. Substantial IT investment also resulted in lower stock returns during the distressing episodes in 2023.

While digital banking helps banks attract deposits during economic upturns, it has both positive and negative consequences, as tech-savvy depositors are more inclined to swiftly move their deposits elsewhere during financial turmoil. Consistent with this, we observe that while the average bank experienced a 27.5 percent surge in webpage traffic in March 2023, banks with lower branch density experienced a notably higher increase in online traffic during that period. This shift in online traffic also negatively and significantly predicted stock returns around the collapses of Silicon Valley Bank and First Republic Bank. These results align with both stock returns and increases in online banking traffic being linked to deposit instability.

Our discoveries emphasize the significance of branch density and its impact on deposit stability. Digital banking allows banks with lower branch density to draw in deposits and expand their funding capacity. Nevertheless, digital banking and low branch density diminish the value of the bank-depositor relationship, transferring the depositor base to corporations and tech-savvy individuals with substantial, mainly uninsured deposits. These changes in the bank’s depositor composition prove detrimental during market downturns, as digital banks with lower branch density undergo larger deposit outflows and experience poorer stock performance.

Although increasing account ownership among the unbanked is an important initial step in broadening financial inclusion, mere ownership doesn’t ensure that individuals reap the benefits of financial access. It’s equally vital for them to confidently and effectively use their accounts. Unfortunately, many unbanked and underbanked adults lack the necessary education or exposure to navigate financial systems, resulting in uncertainty regarding the advantages of financial access.

So, how can policymakers and advocates maximize the advantages of financial access? A recent paper by the Global Findex team delves into the role of financial education in enhancing both the adoption and usage of financial accounts. The paper investigates the demographics of the unbanked and underbanked, their barriers to account ownership, and the potential impact of improving financial knowledge for these groups. This overview aims to shed light on their findings.

The unbanked are more likely to be women, impoverished, and have lower educational attainment. In various visits to countries in Africa and Asia, the Global Findex team conducted qualitative surveys to evaluate individuals’ ability to conduct financial transactions independently. Many respondents, particularly women receiving government payments, revealed a dependency on family or friends for financial transactions. Those seeking assistance from bank agents often faced additional fees or unofficial payments. In some regions, mobile money users reported receiving fraudulent calls requesting their account PIN codes, leaving them with limited recourse if defrauded.

Although qualitative, these findings are consistent with data from the Global Findex 2021, which revealed that 1.4 billion adults, constituting 24 percent of the global adult population, remain unbanked, despite a 50 percent increase in account ownership over the decade.

Most unbanked adults are women, belong to the bottom 40% of income-earners, and possess only primary education or less. These factors collectively highlight the necessity for user-friendly product design and robust consumer safeguards to ensure that the unbanked benefit from financial access.

The Global Findex team discovered that 64 percent of unbanked adults would not be able to manage an account without assistance. In some countries, such as Pakistan and South Sudan, this figure rises to as high as 80 percent. Unbanked women in developing countries are 10 percentage points more likely than men to express the need for assistance.

Enhanced financial literacy and capability have the potential to inspire and enable the secure and beneficial utilization of financial services, not just for unbanked individuals but also for some banked adults. In Sub-Saharan Africa, one-third of mobile money account holders rely on assistance from family or agents to operate their accounts. This dependence may render some banked adults more susceptible to exploitation or misunderstanding account rules, as observed with one in five wage earners in developing countries facing unexpected fees on their financial institution accounts.

Moreover, there are 430 million inactive account holders worldwide, termed as the underbanked, who possess accounts but do not actively use them, thereby limiting the benefits they could derive. Most of these inactive account holders reside in India, and reasons for their inactivity include the distance to financial institutions, lack of trust in financial systems, and the absence of a perceived need for an account. This underlines that mere possession of an account is insufficient to encourage active use.

How can we tackle the issue of financial confidence for banked, unbanked, and underbanked individuals?

Lack of adequate financial education and consumer safeguards could leave low-income individuals without the necessary experience, unable to fully leverage the advantages of owning an account, and vulnerable to fraud, discrimination, and financial exploitation. Financial education plays a significant role in mitigating some of these potential negative consequences, but consumer safeguards are equally crucial for ensuring customer safety and bolstering confidence in the financial system.

Exercises such as ‘mystery shopping’ have revealed that inexperienced consumers received inadequate information on account costs and were seldom presented with the most cost-effective products. This occurs due to staff incentives to sell more expensive products, placing consumers at risk. Financial institutions need to align employee incentives with consumer protection requirements to mitigate this risk. Additionally, presenting clear and concise information about credit and savings products can assist consumers in making sound financial decisions. For example, a study in Mexico and Peru found that clear and concise information led to consumers making more appropriate choices in financial products. Allowing users to monitor their account balances in real-time via phone or the internet also builds trust and promotes account usage.

Regulators also play a pivotal role in creating an environment conducive to consumers by addressing critical information gaps and mandating financial institutions to furnish complete information about available account options. By ensuring transparency from financial institutions, regulators can empower individuals to make informed decisions and enhance financial inclusion.

Our recent research paper addresses this question by examining a significant program implemented in Brazil. In March 2012, the Brazilian government announced an increase in credit supply through two of its largest commercial banks, collectively responsible for 38 percent of Brazil’s outstanding credit prior to the intervention. Focusing on term loans to firms, we observed a substantial rise in lending by public banks, offering significantly lower interest rates compared to private banks. While there were advantages such as reduced loan interest rates and positive employment effects, we also noted substantial costs, including increased corporate debt and higher loan default rates for state-owned banks.

Conducting a comprehensive analysis of government interventions in credit markets is challenging. Decisions to intervene, adjustments made by private banks in their loan interest rates, and outcomes for beneficiaries can be linked to economic characteristics that are not easily observable, which may affect the analysis. However, our study overcomes these challenges by leveraging rich data and a unique setting outside of a financial crisis. When assessing the effects of the policy on loan interest rates, firm leverage, and loan default, we focused on loans and firms with identical characteristics such as sectors, location, and maturity. This allowed us to specifically pinpoint changes resulting from the policy’s introduction in early 2012.

We found that private banks notably reduced their interest rates in response to the increase in lending by state-owned banks, as depicted in Figure 1. The increase in lending also resulted in higher leverage for firms borrowing from public banks compared to those dealing exclusively with private banks. The difference in interest rates and increased leverage can both influence firm loan default risk. Figure 2 illustrates the delinquency rates for public and private banks, showing that public banks experienced a deterioration in loan portfolio quality following the policy announcement. Notably, this difference was observed primarily among leveraged firms, indicating increased risk for public banks due to higher leverage.

Additionally, our analysis revealed that firms with access to public banks experienced increased employment after the policy. However, to ensure unbiased results, we compared employment and output in neighboring cities with varying exposure levels to public banks. This regional analysis indicated that the credit expansion associated with state-owned banks correlated with amplified output and employment growth in cities more exposed to these banks. Despite this, the impact on output growth in relation to credit expansion was relatively modest compared to similar studies on credit supply’s real effects.

To summarize, our study delves into the costs and benefits of interventions using lending from state-owned banks. While there are potential benefits, like increased competition and lower interest rates by private banks, the real effects are moderate and involve higher firm leverage and default. Our findings stress that policymakers should be mindful of the costs associated with increased leverage and default resulting from public lending programs outside of financial crises, even when offering lower loan interest rates compared to prevailing market rates.